As people across the globe are becoming more interested in watching their weight, eating more healthily, and avoiding obesity, a system that can measure calories and nutrition in everyday meals can be very useful. Recently, there has been an increase in the usage of personal mobile technology such as smartphones or tablets, which users carry with them practically all the time. In this paper, we proposed a food calorie and nutrition measurement system that can help patients and dieticians to measure and manage daily food intake. Our system is built on food image processing and uses nutritional fact tables. Via a special calibration technique, our system uses the built-in camera of such mobile devices and records a photo of the food before and after eating it in order to measure the consumption of calorie and nutrient components. The proposed algorithm used color, texture and contour segmentation and extracted important features such as shape, color, size and texture. Using various combinations of these features and applying a support vector machine as a classifier, a good classification was achieved and simulation results show that the algorithm recognizes food categories with an accuracy rate of 92.2%, on average.

Several models of visual word recognition suggest a processing hierarchy; basic orthographic features are processed early and whole-word representations are processed late in the hierarchy. Unfortunately, given the extreme efficiency of the visual word recognition system, studies typically focus on one specific level of the processing hierarchy (e.g., orthographic, phonological and/or semantic processing). Furthermore, different paradigms are used to study different levels of the hierarchy. Fortunately, data across different studies in the literature do converge to two distinct temporal thresholds for letter perception and whole-word integration. The current experiments assessed the temporal thresholds for both letter perception and whole-word integration using a single novel paradigm. The results demonstrated distinct temporal thresholds for letter perception and whole-word integration which agree with those reported in the literature. Thus, the current experiments provide further behavioral evidence that the visual word recognition is a hierarchical process.

Several models of visual word recognition suggest a processing hierarchy; basic orthographic features are processed early and whole-word representations are processed late in the hierarchy. Unfortunately, given the extreme efficiency of the visual word recognition system, studies typically focus on one specific level of the processing hierarchy (e.g., orthographic, phonological and/or semantic processing). Furthermore, different paradigms are used to study different levels of the hierarchy. Fortunately, data across different studies in the literature do converge to two distinct temporal thresholds for letter perception and whole-word integration. The current experiments assessed the temporal thresholds for both letter perception and whole-word integration using a single novel paradigm. The results demonstrated distinct temporal thresholds for letter perception and whole-word integration which agree with those reported in the literature. Thus, the current experiments provide further behavioral evidence that the visual word recognition is a hierarchical process.

The area of speech recognition by machine is one of the most popular and complicated subjects in the current multimedia field. Linear predictive coding (LPC) is a useful technique for voice coding in speech analysis and synthesis. The first objective of this research was to establish a prototype of the residual-excited linear predictive (RELP) vocoder system in a real-time environment. Although its transmission rate is higher, the quality of synthesized speech of the RELP vocoder is superior to that of other vocoders. As well, it is rather simple and robust to implement. The RELP vocoder uses residual signals as excitation rather than periodic pulse or white noise. The RELP vocoder was implemented with Texas Instruments TMS320C6711 DSP starter kit (DSK) using C. Identifying vowel sounds is an important element in recognizing speech contents. The second objective of research was to explore a method of characterizing vowels by means of parameters extracted by the RELP vocoder, which was not known to have been used in speech recognition, previously. Five English vowels were chosen for the experimental sample. Utterances of individual vowel sounds and of the vowel sounds in one-syllable-words were recorded and saved as WAVE files. A large sample of 20-ms vowel segments was obtained from these utterances. The presented method utilized 20 samples of a segment's frequency response, taken equally in logarithmic scale, as a LPC frequency response vector. The average of each vowel's vectors was calculated. The Euclidian distances between the average vectors of the five vowels and an unknown vector were compared to classify the unknown vector into a certain vowel group. The results indicate that, when a vowel is uttered alone, the distance to its average vector is smaller than to the other vowels' average vectors. By examining a given vowel frequency response against all known vowels' average vectors, individually, one can determine to which vowel group the given vowel belongs. When a vowel is uttered with consonants, however, variances and covariances increase. In some cases, distinct differences may not be recognized among the distances to a vowel's own average vector and the distances to the other vowels' average vectors. Overall, the results of vowel characterization did indicate an ability of the RELP vocoder to identify and classify single vowel sounds.

The current education systems in elementary schools are usually using traditional teaching methods such as paper and pencil or drawing on the board. The benefit of paper and pencil is their ease of use. Researchers have tried to bring this ease of use to computer-based educational systems through the use of sketch-recognition. Sketch-recognition allows students to draw naturally while at the same time receiving automated assistance and feedback from the computer. There are many sketch-based educational systems for children. However, current sketch-based educational systems use the same sketch recognizer for both adults and children. The problem of this approach is that the recognizers are trained by using sample data drawn by adults, even though the drawing patterns of children and adults are markedly different. We propose that if we make a separate recognizer for children, we can increase the recognition accuracy of shapes drawn by children. By creating a separate recognizer for children, we improved the recognition accuracy of children’s drawings from 81.25% (using the adults’ threshold) to 83.75% (using adjusted threshold for children). Additionally, we were able to automatically distinguish children’s drawings from adults’ drawings. We correctly identified the drawer’s age (age 3, 4, 7, or adult) with 78.3%. When distinguishing toddlers (age 3 and 4) from matures (age 7 and adult), we got a precision of 95.2% using 10-fold cross validation. When we removed adults and distinguished between toddlers and 7 year olds, we got a precision of 90.2%. Distinguishing between 3, 4, and 7 year olds, we got a precision of 86.8%. Furthermore, we revealed that there is a potential gender difference since our recognizer was more accurately able to recognize the drawings of female children (91.4%) than the male children (85.4%). Finally, this paper introduces a sketch-based teaching assistant tool for children, EasySketch, which teaches children how to draw digits and characters. Children can learn how to draw digits and characters by instructions and feedback.

The area of speech recognition by machine is one of the most popular and complicated subjects in the current multimedia field. Linear predictive coding (LPC) is a useful technique for voice coding in speech analysis and synthesis. The first objective of this research was to establish a prototype of the residual-excited linear predictive (RELP) vocoder system in a real-time environment. Although its transmission rate is higher, the quality of synthesized speech of the RELP vocoder is superior to that of other vocoders. As well, it is rather simple and robust to implement. The RELP vocoder uses residual signals as excitation rather than periodic pulse or white noise. The RELP vocoder was implemented with Texas Instruments TMS320C6711 DSP starter kit (DSK) using C. Identifying vowel sounds is an important element in recognizing speech contents. The second objective of research was to explore a method of characterizing vowels by means of parameters extracted by the RELP vocoder, which was not known to have been used in speech recognition, previously. Five English vowels were chosen for the experimental sample. Utterances of individual vowel sounds and of the vowel sounds in one-syllable-words were recorded and saved as WAVE files. A large sample of 20-ms vowel segments was obtained from these utterances. The presented method utilized 20 samples of a segment's frequency response, taken equally in logarithmic scale, as a LPC frequency response vector. The average of each vowel's vectors was calculated. The Euclidian distances between the average vectors of the five vowels and an unknown vector were compared to classify the unknown vector into a certain vowel group. The results indicate that, when a vowel is uttered alone, the distance to its average vector is smaller than to the other vowels' average vectors. By examining a given vowel frequency response against all known vowels' average vectors, individually, one can determine to which vowel group the given vowel belongs. When a vowel is uttered with consonants, however, variances and covariances increase. In some cases, distinct differences may not be recognized among the distances to a vowel's own average vector and the distances to the other vowels' average vectors. Overall, the results of vowel characterization did indicate an ability of the RELP vocoder to identify and classify single vowel sounds.

Image-based face recognition has attained wide applications during the past decades in commerce and law enforcement areas, such as mug shot database matching, identity authentication, and access control. Existing face recognition techniques (e.g., Eigenface, Fisherface, and Elastic Bunch Graph Matching, etc.), however, do not perform well when the following case inevitably exists. The case is that, due to some variations in imaging conditions, e.g., pose and illumination changes, face images of the same person often have different appearances. These variations make face recognition techniques much challenging. With this concern in mind, the objective of my research is to develop robust face recognition techniques against variations. This thesis addresses two main variation problems in face recognition, i.e., pose and illumination variations. To improve the performance of face recognition systems, the following methods are proposed: (1) a face feature extraction and representation method using non-uniformly selected Gabor convolution features, (2) an illumination normalization method using adaptive region-based image enhancement for face recognition under variable illumination conditions, (3) an eye detection method in gray-scale face images under various illumination conditions, and (4) a virtual pose generation method for pose-invariant face recognition. The details of these proposed methods are explained in this thesis. In addition, we conduct a comprehensive survey of the existing face recognition methods. Future research directions are pointed out.

Recognition and tracking of multiple objects in video sequences is one of the main challenges in computer vision that currently deserves a lot of attention from researchers. Almost all the reported approaches are very application-dependent and there is a lack of a general methodology for dynamic object recognition and tracking that can be instantiated in particular cases. In this thesis, the work is oriented towards the definition and development of such a methodology which integrates object recognition and tracking from a general perspective using a probabilistic framework called PIORT (probabilistic integrated object recognition and tracking framework). It include some modules for which a variety of techniques and methods can be applied. Some of them are well-known but other methods have been designed, implemented and tested during the development of this thesis.The first step in the proposed framework is a static recognition module that provides class probabilities for each pixel of the image from a set of local features. These probabilities are updated dynamically and supplied to a tracking decision module capable of handling full and partial occlusions. The two specific methods presented use RGB colour features and differ in the classifier implemented: one is a Bayesian method based on maximum likelihood and the other one is based on a neural network. The experimental results obtained have shown that, on one hand, the neural net based approach performs similarly and sometimes better than the Bayesian approach when they are integrated within the tracking framework. And on the other hand, our PIORT methods have achieved better results when compared to other published tracking methods. All these methods have been tested experimentally in several test video sequences taken with still and moving cameras and including full and partial occlusions of the tracked object in indoor and outdoor scenarios in a variety of cases with different levels of task complexity. This allowed the evaluation of the general methodology and the alternative methods that compose these modules.A Probabilistic Integrated Object Recognition and Tracking Framework for Video Sequences / El reconocimiento y seguimiento de múltiples objetos en secuencias de vídeo es uno de los principales desafíos en visión por ordenador que actualmente merece mucha atención de los investigadores. Casi todos los enfoques reportados son muy dependientes de la aplicación y hay carencia de una metodología general para el reconocimiento y seguimiento dinámico de objetos, que pueda ser instanciada en casos particulares. En esta tesis, el trabajo esta orientado hacia la definición y desarrollo de tal metodología, la cual integra reconocimiento y seguimiento de objetos desde una perspectiva general usando un marco probabilístico de trabajo llamado PIORT (Probabilistic Integrated Object Recognition and Tracking). Este incluye algunos módulos para los que se puede aplicar una variedad de técnicas y métodos. Algunos de ellos son bien conocidos, pero otros métodos han sido diseñados, implementados y probados durante el desarrollo de esta tesis.El primer paso en el marco de trabajo propuesto es un módulo estático de reconocimiento que provee probabilidades de clase para cada píxel de la imagen desde un conjunto de características locales. Estas probabilidades son actualizadas dinámicamente y suministradas a un modulo decisión de seguimiento capaz de manejar oclusiones parciales o totales. Se presenta dos métodos específicos usando características de color RGB pero diferentes en la implementación del clasificador: uno es un método Bayesiano basado en la máxima verosimilitud y el otro método está basado en una red neuronal. Los resultados experimentales obtenidos han mostrado que, por una parte, el enfoque basado en la red neuronal funciona similarmente y algunas veces mejor que el enfoque bayesiano cuando son integrados dentro del marco probabilístico de seguimiento. Por otra parte, nuestro método PIORT ha alcanzado mejores resultados comparando con otros métodos de seguimiento publicados. Todos estos métodos han sido probados experimentalmente en varias secuencias de vídeo tomadas con cámaras fijas y móviles incluyendo oclusiones parciales y totales del objeto a seguir, en ambientes interiores y exteriores, en diferentes tareas y niveles de complejidad. Esto ha permitido evaluar tanto la metodología general como los métodos alternativos que componen sus módulos.

This study investigated the associations between emotion recognition ability and autistic traits in a sample of non-clinical young adults. Two hundred and forty nine individuals took part in an emotion recognition test, which assessed recognition of 12 emotions portrayed by actors. Emotion portrayals were presented as short video clips, both with and without sound, and as sound only. Autistic traits were assessed using the Autism Spectrum Quotient (ASQ) questionnaire. Results showed that men had higher ASQ scores than women, and some sex differences in emotion recognition were also observed. The main finding was that autistic traits were correlated with several measures of emotion recognition. More specifically, ASQ-scores were negatively correlated with recognition of fear and with recognition of ambiguous stimuli.

It is difficult for people to recognize currencies from different countries. Our aim is to help people solve this problem. However, currency recognition systems that are based on image analysis entirely are not sufficient. Our system is based on image processing and makes the process automatic and robust. We use SEK and Chinese RMB as examples to illustrate the technique. Color and shape information are used in our algorithm.